1. Technical Field
The present invention relates to drivers, electronic devices, and the like.
2. Related Art
Display devices (liquid-crystal display devices, for example) are used in a variety of electronic devices, including projectors, information processing apparatuses, mobile information terminals, and the like. Increases in the resolutions of such display devices continue to progress, and as a result, the time a driver drives a single pixel is becoming shorter. For example, phase expansion driving is used as a method for driving an electro-optical panel (a liquid-crystal display panel, for example). According to this driving method, for example, eight source lines are driven at one time, and the process is repeated 160 times to drive 1,280 source lines. In the case where a WXGA (1,280×768 pixels) panel is to be driven, the stated 160 instances of driving (that is, the driving of a single horizontal scanning line) is thus repeated 768 times. Assuming a refresh rate of 60 Hz, a simple calculation shows that the driving time for a single pixel is approximately 135 nanoseconds. In actuality, there are periods where pixels are not driven (blanking intervals and the like, for example), and thus the driving time for a single pixel becomes even shorter, at approximately 70 nanoseconds.
Past drivers for driving such electro-optical panels have included D/A conversion circuits for converting tone data (image data) of each pixel into data voltages and amplifier circuits that drive the pixels with the data voltages. This is done in order for the amplifier circuits to carry out impedance conversion and supply charges for capacitance on the electro-optical panel side (parasitic capacitance of interconnects, pixel capacitance, and the like, for example). In other words, past drivers have been configured to supply required charges corresponding to the data voltages.
However, with the increases in resolutions of electro-optical panel as mentioned above, it is becoming difficult for the amplifier circuits to finish writing the data voltages within the required time. For example, in the above WXGA example, it is necessary for the writing for a single pixel to finish within 70 nanoseconds, and thus the write time becomes even shorter if an attempt to further increase the resolution is made. For the amplifier circuits to drive the pixels at high speeds, it is necessary to have a wide output range corresponding to the range of the data voltages, and to be able to supply the charges at a high speed at any voltage within that output range. Achieving both requires, for example, an increase in the bias voltage of the amplifier circuits, resulting in a further increase in power consumption in drivers as increases in resolution progress.
A method that drives an electro-optical panel through capacitor charge redistribution (called “capacitive driving” hereinafter) can be considered as a driving method for solving such problems. For example, JP-A-2000-341125 and JP-A-2001-156641 disclose techniques that use capacitor charge redistribution in D/A conversion. In a D/A conversion circuit, both driving-side capacitance and load-side capacitance are included in an IC, and charge redistribution occurs between those capacitances. For example, assume such a load-side capacitance of the D/A conversion circuit is replaced with the capacitance of the electro-optical panel external to the IC and used as a driver. In this case, charge redistribution occurs between the driver-side capacitance and the electro-optical panel-side capacitance.
However, although charges can be freely supplied with an amplifier circuit, capacitive driving uses charge redistribution, and thus there is a problem that capacitive driving suffers from a drop in data voltage accuracy. For example, in capacitive driving, data voltages are determined by capacitance ratios, but because an electro-optical panel-side capacitance is a capacitance external to the driver IC, it is more difficult to set the capacitance ratio exactly than in the case of a capacitance internal to the IC. Alternatively, there are cases where charge conservation breaks down due to operations within the electro-optical panel and the like (connections between data lines and source lines, for example) and leads to data voltage errors.